Problem running Boinc on T2000 and Solaris 10 !
Questions and Answers :
Unix/Linux :
Problem running Boinc on T2000 and Solaris 10 !
Message board moderation
| Author | Message |
|---|---|
|
Didn? Send message Joined: 8 Mar 06 Posts: 4 Credit: 357,584 RAC: 0 
|
Hi, can´t run Boinc on a SUN T2000 with Solaris 10 01/06, get the following message and the cpu benchmarks timed out: ACTIVE_TASK_SET::check_app_exited(): pid 1490 not found on my SUN 420R with 4 cpu´s Boinc is running without problems. thx for any idea...Didn? ID: 260042 ·  |
 Jim-R.  Send message Joined: 7 Feb 06 Posts: 1494 Credit: 194,148 RAC: 0 
|
The message you are receiving means that a task was told to exit but it didn't exit as it was expected to but when it was tried to be used again it wasn't found in the active task list. From the way you describe it sounds probable that something in the benchmark routine failed. If you have boinc set up to start from a menu or an icon on your desktop or something I suggest running it from a terminal to see if there are any other errors listed such as missing libraries etc. Jim Jim Some people plan their life out and look back at the wealth they've had. Others live life day by day and look back at the wealth of experiences and enjoyment they've had. ID: 260045 · |
|
Didn? Send message Joined: 8 Mar 06 Posts: 4 Credit: 357,584 RAC: 0
|
The message you are receiving means that a task was told to exit but it didn't exit as it was expected to but when it was tried to be used again it wasn't found in the active task list. From the way you describe it sounds probable that something in the benchmark routine failed. If you have boinc set up to start from a menu or an icon on your desktop or something I suggest running it from a terminal to see if there are any other errors listed such as missing libraries etc. Hello Jim, thank for your answer, i only work on the shell! i tryed a client for specially Sparc IV and the messages are simular and still after the benchmark nothing more happens :-( : 2006-03-10 14:01:08 [---] Starting BOINC client version 4.43 for sparc-sun-solaris2.10 2006-03-10 14:01:08 [---] Data directory: /export/home/dh 2006-03-10 14:01:08 [http://setiathome.berkeley.edu/] Computer ID: not assigned yet; location: ; project prefs: default 2006-03-10 14:01:08 [---] No general preferences found - using BOINC defaults 2006-03-10 14:01:08 [---] Remote control not allowed; using loopback address 2006-03-10 14:01:10 [---] Running CPU benchmarks 2006-03-10 14:06:08 [---] CPU benchmarks timed out, using default values 2006-03-10 14:06:08 [---] CPU benchmarks timed out, using default values 2006-03-10 14:06:08 [---] Resuming computation and network activity 2006-03-10 14:06:08 [---] ACTIVE_TASK_SET::check_app_exited(): pid 1858 not found 2006-03-10 14:06:08 [---] ACTIVE_TASK_SET::check_app_exited(): pid 1858 not found 2006-03-10 14:06:08 [---] Insufficient work; requesting more 2006-03-10 14:06:08 [---] Suspending computation and network activity - running CPU benchmarks 2006-03-10 14:06:10 [---] Running CPU benchmarks any idea? Didn? ID: 260116 · |
|
Jim-R. Send message Joined: 7 Feb 06 Posts: 1494 Credit: 194,148 RAC: 0
|
The message you are receiving means that a task was told to exit but it didn't exit as it was expected to but when it was tried to be used again it wasn't found in the active task list. From the way you describe it sounds probable that something in the benchmark routine failed. If you have boinc set up to start from a menu or an icon on your desktop or something I suggest running it from a terminal to see if there are any other errors listed such as missing libraries etc. I could suggest checking the file beginning with "stderr" in the boinc directory to see if there are any other error messages there. There were no error messages on the terminal after you typed in the command? Another thing I noticed is it seems that you have not attached this computer yet. An output line above says you don't have an id number for this one yet. I see two suns and a windows comp in your account but both of the Suns have computer id's. If it is another computer from those try attaching to the project on the command line when you run it to see if it might need a computer id before it would run the benchmarks. If this doesn't help it seems as some problem is killing the benchmarking routine. I run Linux so I will be of limited help on the technical details of a SunOs but I will help what I can. Jim /edit/ Sorry, I reread your post and I do see it as a different computer T2000 instead of Ultra 80 and UltraSparc (don't remember the rest of it). Jim Some people plan their life out and look back at the wealth they've had. Others live life day by day and look back at the wealth of experiences and enjoyment they've had. ID: 260130 · |
|
Didn? Send message Joined: 8 Mar 06 Posts: 4 Credit: 357,584 RAC: 0
|
The message you are receiving means that a task was told to exit but it didn't exit as it was expected to but when it was tried to be used again it wasn't found in the active task list. From the way you describe it sounds probable that something in the benchmark routine failed. If you have boinc set up to start from a menu or an icon on your desktop or something I suggest running it from a terminal to see if there are any other errors listed such as missing libraries etc. Hi Jim, i have no "stderr", all i have is: # find . . ./boinc ./boincmgr ./binstall.sh ./run_client ./lockfile ./projects ./projects/setiathome.berkeley.edu ./client_state.xml ./account_setiathome.berkeley.edu.xml ./client_state_prev.xml i start the client with: ./boinc -attach_project http://setiathome.berkeley.edu KEY There are no other error messages :-( ID: 260140 · |
|
Jim-R. Send message Joined: 7 Feb 06 Posts: 1494 Credit: 194,148 RAC: 0
|
Ok. Were you running boinc at the time that the above list was generated? If not, delete the lockfile and try again. If that is a stale lockfile left behind from it exiting ungraciously that may be the reason. I noticed that in both of your former posts it gave the exact same process id. That shouldn't happen unless it's picking it up from something left behind. I don't know where you would look for system errors on the sunos but you might can check them. You might try viewing the contents of the lockfile too. I checked mine and it gives some messages from Einstein but I didn't see any for Seti. I have rechecked all the info I can find on this error message and still can only find the wiki info that I posted earlier. Still no actual "fix". I hate to do it but since you haven't even got it to attach yet I suggest deleting everything and starting over. It is possible that one of the files could be corrupt. Either that or keep bumping this post up (or create a new one) and try to get someone that is more familiar with SunOs in here. You might could also try posting in the "number crunching" forum and prefix your post with @Eric. He is a lead developer for the Seti application and may be able to help, and you may also want to post in one of the boinc forums. I will keep trying to find something but right now it doesn't look promising. Jim Jim Some people plan their life out and look back at the wealth they've had. Others live life day by day and look back at the wealth of experiences and enjoyment they've had. ID: 260179 · |
Dotsch  Send message Joined: 9 Jun 99 Posts: 2422 Credit: 919,393 RAC: 0
|
You can not have a stderr.txt or can attach to the project, because boinc client hang on the benchmark, and could not start the app. Do you have running boinc on a NFS filesystem ? - If yes work it on a local ufs filesystem ? There was some posts here, that sometimes the benchmark hang. The error was not clearly reproducable. But for one user helped to reinstall or reattach to the project. So far a other on has rebootet the system. Please, can mail me (seti_boinc@dotsch.de) the output from the command "truss ./boinc_client > out 2>&1". - But let it run some time (about five minutes). ID: 260191 · |
|
SunMicrosystemsLLG Send message Joined: 4 Jul 05 Posts: 102 Credit: 1,360,617 RAC: 0 
|
Hi, We also had some problems trying to get BOINC to benchmark on T2000s. Since the UltraSPARC T1 CPU has upto 8 physical cores and 4 threads executing per core; 32 'virtual CPUs' are presented to Solaris. This means the BOINC application tries to lauch 32 instances of the benchmarking process. Unfortunatley, the US T1 only has 1 FPU shared between the 8 physical cores, so all 32 benchmarking processes are fighting over 1 floating point unit. The only way around it that I got to work was to set up a zone on the system and add a processor set to the zone which only contained a single CPU - This way the BOINC app would only launch one benchmarking process and there would be no contention at the FPU. Once the benchmarking process had finished I could then add more CPUs to the processor set - be careful not to add too many, I added CPUs 0, 4, 8 & 12 (virtual CPUs from different cores) so that each vCPU assigned to SETI had it's own L1 cache. I didn't get as far as assigning 1 vCPU from each core, so I'm not sure how performance degrades as you increase the vCPU count. Hope this helps. 
ID: 261630 · |
|
Didn? Send message Joined: 8 Mar 06 Posts: 4 Credit: 357,584 RAC: 0
|
Thanks to all helping to solve this "problem" ;-) Didn? ID: 261893 · |
©2025 University of California
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.